| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
Servicio de Endocrinología (R.A., F.S.-F.), Hospital Carlos III-C.I.C., Instituto de Salud Carlos III, Madrid 28029, Spain; and Servicio de Endocrinología (L.C., C.N.), Hospital Ramón y Cajal, Madrid 28034, Spain
Address all correspondence and requests for reprints to: Dr. Franco Sánchez-Franco, Hospital Carlos III. C/Sinesio Delgado, 10-12, Madrid 28029. E-mail: fsanchez{at}hciii.insalud.es.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GH, usually considered to be of anterior pituitary origin, has been detected within the rodent CNS. Immunoassayable GH can be found in the fetal whole brain as early as d 10 of gestation before its detection in the fetal pituitary (2). Brain GH immunoreactivity is not affected by hypophysectomy. GH mRNA is present in some regions of the brain in which GH immunoreactivity has been reported (3).
Several experimental models verify the importance of this hormone in brain development. The Snell dwarf mouse with loss-of-function mutations at Pit1 and the GHRH receptor-deficient little mouse exhibit microcephalic cerebrum with hypomyelination, retarded neuronal growth with poor synaptogenesis, and reduced levels of spontaneous locomotion activity (4, 5). The hypomyelination was found to be due to arrested glial proliferation, suggesting that the action of GH on the proliferation and maturation of both glial and neuronal cells is a necessary precondition for myelin formation. After the administration of bovine GH to Snell dwarf mice during the first 20 d of postnatal life, the retarded neuronal growth was restored to normal (6). In Laron syndrome, because of a mutant GH receptor (GHR) defect, slow motor development, a small cranium and impairment of intellectual development are observed (7, 8). Transgenic mice overexpressing bovine, ovine, or rat GH exhibit enhancement of growth, with increased spinal cord and brain weight as well as a number of endocrine abnormalities (9, 10). In recent years the actions that GH may exert in the brain have received a lot of attention in human adults with GH deficiency. These include the effects of the hormone on appetite, cognitive functions, energy, memory, mood, neuroprotection, sleep, and well-being (11).
Specific receptors for GH have been localized in discrete areas of the rat CNS during development. An increase in GHR/binding protein immunoreactivity is evident from 12- to 18-d-old embryos with localization in the nervous system beginning in the neuroepithelium (12). The presence of GH and its receptor during CNS development support the concept of a GH axis in the brain and suggest that this hormone may play a role in neurogenesis and gliogenesis.
During embryonic development, discrete populations of proliferating progenitor cells in the neural tube respond to signals that determine their commitment to differentiate into neuronal or glial lineage (13). It is well established that IGF-I influences both early and late events in CNS neurogenesis and gliogenesis (14, 15); however, it remains unclear whether GH can influence these processes.
Among all the effects exerted by GH on brain function, some may result from a direct action of the hormone on its brain receptor, whereas other effects may be due to GH-induced peripheral mediators, such as IGF-I that can gain access to the CNS. Interestingly, despite the fundamental role of GH in postnatal growth and development and the importance in the regulation of its target gene IGF-I, the role of GH on the tissue specific regulation of IGF-I gene expression during development is unclear (16). Because there is evidence for GH synthesis within the brain and it has been demonstrated that the GH receptor distribution correlates with IGF-I expression, there is possibility that an endogenous GH/IGF-I/IGF-binding protein-3 (IGFBP-3) system is involved in brain growth and maturation (12, 16, 17).
IGF-I plays a fundamental role in growth and development, both as a mediator of many of the actions of GH and as a locally acting mitotrophic factor promoting tissue growth and cell differentiation (16). Increasing evidence supports a role for IGF-I in CNS development. The expression pattern of IGF-I system proteins during brain growth suggests highly regulated and developmentally timed IGF actions on specific neural cell populations (18, 19). IGF-I expression in the developing rodent brain begins by at least embryonic d 14 and occurs predominantly in neuronal cells. Analysis of IGF-I mRNA indicates that the peak of IGF-I expression in brain occurs during the second week of postnatal life (20, 21). The type I IGF receptor, which mediates IGF-I signals, is expressed in brain as early as embryonic d 13 and has been found to be most abundant at embryonic d 15 and 20 (22).
Because the role of GH in fetal CNS maturation is poorly understood, in this study we aim to investigate the neurotrophic action of GH during brain development by the analysis of its proliferative and differentiative actions on fetal cerebral cortical cells in primary mixed cultures.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-32P] were from Amersham Pharmacia Biotech (Amersham, Buckinghamshire, UK). Poly-L-ornithine was from Sigma.
Buffers and media
Fetal bovine serum (FBS), horse serum (HS), Hanks’ balanced salt solution (HBSS), PBS, DMEM, and antibiotics were from BioWhittaker, Inc. (Walkersville, MD). Neurobasal medium, N2, and B27 supplements were from Life Technologies, Inc. (Gaithersburg, MD). L-15 was from ICN (Costa Mesa, CA).
Defined medium consisted of DMEM (glucose 1 g/liter): (1:1) Ham’s F12, supplemented with NaHCO3 (1.2 g/liter), glucose (6 g/liter) (Merck, Darmstadt, Germany), transferrin (0.1 mg/ml) (Roche Molecular Biochemicals, Mannheim, Germany), putrescine (10-5 M) and sodium selenite (2 x 10-8 M), corticosterone (10-7 M), HEPES (15 mM), T3 (10-10 M) (Sigma), L-glutamine (4 mM), and penicillin-streptomycin (100 U/ml) (BioWhittaker, Inc.).
Primary cell cultures
Preparation of primary dispersed cell cultures of fetal rat cerebral cortex was done as previously described (23). Timed pregnant Wistar rats were raised in our laboratory. The embryos were removed from the isoflurane-anesthetized mothers. The cerebral cortices were dissected under sterile conditions and the cells mechanically dispersed.
Animal care was conducted in accordance with the guidelines established by the Real Decreto 223/March 14, 1988, and Orden October 13, 1989, on the protection of animals used in scientific research.
Cultures of 14-d-old embryos (E14).
The cortices from E14 rats were dissected in HBSS 6.5 g/liter glucose and then transferred to culture medium (glutamine-free L-15 supplemented with 1% N2, 30 mM glucose, 25.8 mM NaHCO3, and penicillin-streptomycin (100 UI/ml) for 24 h. Then the medium was replaced with neurobasal medium supplemented with 2% B27 and incubated for variable times. Cells were plated onto poly-L-ornithine-coated glass coverslips (24 wells) at a density of 7.5 x 104 cells/well. The medium was changed on the third day in vitro (DIV).
Cultures of 17-d-old embryos (E17).
The cortices from E17 rats were dissected in HBSS 6.5 g/liter glucose and then were suspended in DMEM supplemented with 2.5% FBS and 2.5% HS for 2 h (proliferation experiments) or DMEM supplemented with 7.5% HS and 7.5% FBS for 5 h (differentiation experiments). Then the medium was replaced by defined medium for studies performed with cells plated onto poly-L-ornithine-coated 35-mm tissue culture dishes and seeded at a density of 2.5 x 106 cells/35-mm plate. For immunocytochemical studies, the medium was replaced by neurobasal medium supplemented with 2% B27, cells plated onto poly-L-ornithine-coated 24-well culture dishes, and seeded at a density of 7.5 x 104 cells/well.
Experimental design
Incubation with GH
Proliferation experiments.
For immunocytochemical studies, E14 cells were cultured as indicated above and incubated with rhGH (50 ng/ml) for 2, 3, and 7 DIV.
Total proliferation studies in cells from E17 were performed on poly-L-ornithine-coated 35-mm tissue culture dishes. Cells were cultured as indicated above and incubated with rhGH (0.5, 5, 50, and 500 ng/ml) for 24, 48, and 72 h. For immunocytochemical studies, E17 cells were cultured as indicated above and incubated in the presence of rhGH (5 and 50 ng/ml).
Differentiation experiments.
For Western blot studies, E17 cells plated onto poly-L-ornithine-coated 35-mm tissue culture dishes were cultured as indicated above and subsequently incubated with rhGH (5, 50, and 500 ng/ml) and incubated for 48 h. For immunocytochemical studies, E17 cells were plated onto poly-L-ornithine-coated 24-well dishes and cultured as described above. Then cells were incubated for 48 h in the presence of rhGH (50 ng/ml).
Incubation with GH and IGF-I antiserum
Proliferation experiments.
E17 cells were cultured as indicated, and then medium was removed and replaced with defined medium containing GH (5 ng/ml), IGF-I (10 nM), or purified IgG from IGF-I-antiserum (IGF-I-As) (15 µg/ml); GH (5 ng/ml) + IGF-I-As (15 µg/ml) or IGF-I (10 nM) + IGF-I-As (15 µg/ml). IGF-I-As was incubated with GH or IGF-I for 1 h before adding to the culture plates. Cells were then incubated for 48 h.
Differentiation experiments.
E17 cells were cultured as described, and then medium was removed and replaced with defined medium containing GH (5 ng/ml), IGF-I (10 nM), or purified IgG from IGF-I-As (15 µg/ml); GH (5 ng/ml) + IGF-I-As (15 µg/ml); or IGF-I (10 nM) + IGF-I-As (15 µg/ml). IGF-I-As was incubated with GH or IGF-I for 1 h before being added to the culture plates. Then cells were incubated for 48 h.
All the substances tested were added only once, at the beginning of the experiments. Purified IgG from preimmunized rabbits did not alter cell number.
The specific polyclonal antiserum against rhIGF-I1–70 (coded FL-51089), raised in our laboratory, was also used to visualize endogenous IGF-I in the cultures. The characteristics and specificity of FL-51089 antiserum have been described elsewhere (24).
Proliferation and differentiation studies
Total proliferation was determined in cells from E17 by cell counting, quantitation of proliferating cell nuclear antigen (PCNA), [3H]-thymidine incorporation, and immunocytochemical assessment of bromodeoxyuridine (BrdU) in the nuclei.
For cell counting, fetal cerebral cortical cells were grown in 35-mm dishes. Cells were grown as described above. At the end of the experiments, a cell suspension was prepared in HBSS. Cells were counted on a hemocytometer.
For quantitation of PCNA, fetal cerebral cortical cells were grown as indicated above and exposed to rhGH (5 ng/ml) or vehicle for 48 h. The PCNA levels were determined by Western immunoblotting as described below.
To study incorporation of [3H]-thymidine, fetal cerebral cortical cells were grown as described above and incubated with rhGH (0.5, 5, 50, and 500 ng/ml) or vehicle for 48 h. [3H]-Thymidine (1 mCi) was added concomitantly with GH to each well. The amount of [3H]-thymidine incorporated was then determined in cell lysates by scintillation counting.
To assess incorporation of BrdU, fetal cerebral cortical cells were grown in 24-well culture dishes. Cells were exposed to rhGH (5 ng/ml) for 48 h, and BrdU (10 µM) was added 24 h before the cells were fixed. BrdU incorporation was detected immunochemically, and the proportion of labeled cells was determined as described below.
Differentiation was studied in E17 cell cultures by quantification of glial fibrillary acidic protein (GFAP) and ß-tubulin levels, determined by Western immunoblotting and quantitation of GFAP/BrdU-positive cells and ß-tubulin/BrdU-positive cells by immunocytochemical studies.
Western immunoblots
E17 cells growing in 35-mm dishes were lysed in a buffer containing 125 mM Tris-HCl (pH 6.8), 4% sodium dodecyl sulfate, 15% glycerol, and 10% ß-mercaptoethanol. Total protein extracts (20 µg) were resolved by SDS-PAGE and blotted onto polyvinyl difluoride membrane. After blocking the membranes, immunodetection was performed using monoclonal PCNA antibody (1:10,000 dilution, DAKO Corp.), monoclonal ß-tubulin antibody isotype III (1:1500, Sigma), polyclonal GFAP (1:1500 dilution, clone G-A-5, Sigma) antiserum, polyclonal IGF-IRß (1:1000) antiserum, followed by incubation with a goat peroxidase-conjugated antimouse and antirabbit secondary antibodies (1:2000 dilution) (DAKO Corp.). Immunoreactive bands were visualized using an enhanced chemiluminescence detection system (Amersham Pharmacia).
To determine IGFBP-3, the medium was concentrated 50 times and the same procedure described above was followed. After blocking the membranes, immunodetection was performed using the anti-IGF-I-As FL-51089 (24).
Immunoprecipitation
For detection of phosphorylated IGF-IR, equal amounts of cell lysates were incubated with polyclonal antibody against phosphotyrosine residues (PY20) at 4 C according to the manufacturers’ instructions. Protein A-Sepharose (50 µl) was then added for 1 h at 4 C, followed by three washes with lysis buffer and two with PBS. Beads were treated with Laemmli buffer (0.25 M Tris base, 10% sodium dodecyl sulfate, 50% glycerol, 10 mM EDTA, and Coomassie blue), boiled, and separated by SDS-PAGE followed by Western immunoblotting.
Ribonuclease (RNase) protection assay
Total RNA was extracted using the Chomczynski and Sacchi method (25). The RNase protection assay was performed as previously described (26). In summary, total RNA from pools of five 35-mm dishes were hybridized overnight with approximately 600,000 cpm labeled antisense rat GHR, IGF-I, or IGF-IR riboprobe at 45 C. After hybridization, samples were digested, and protected hybrids were isolated by ethanol precipitation after phenol-chloroform extraction and separated according to size on an 8% polyacrylamide/8 M urea denaturing gel. Gels were exposed to x-ray film (Kodak, Cambridge, UK) at -80 C for 24–36 h. Quantitation of the intensities of the autoradiographic bands corresponding to protected hybrids was done by densitometric scanning using Adobe-Photoshop 2.0 (Macintosh, Cupertino, CA) and NIH Image 1.47 programs. All samples were hybridized at the same time with 18S RNA to correct for the differences in gel loading.
RNA probes
Rat GHR probe (27) was transcribed from a 900-bp BglII fragment of a rat GHR cDNA, subcloned into pT7T3 18 U vector. Rat IGF-I probe was subcloned into a pGEM-3 vector (Promega Corp., Madison, WI). RNA probes were synthesized as previously described (26, 28).
Rat IGF-IR probe was subcloned into pGEM-3 vector (Promega Corp.) and linearized with EcoRI to allow for transcription of antisense IGF-IR RNAs. The antisense transcript contained 40 bases of vector sequence and 264 bases complementary to 15 bases of 5'-untranslated sequence as well as to the region encoding the signal peptide and the first 53 amino acids of the -subunit. Hybridization of this probe to rat IGF-I receptor mRNA, followed by RNase digestion, resulted in a protected band of 265 bases (29). The 18S RNA antisense control template containing an 80-bp insert of a highly conserved region of the human ribosomal RNA gene (nucleotides 715–794) was subcloned into the pTR1 plasmid (Ambion, Inc., Austin, TX).
Immunocytochemistry
Immunocytochemistry was performed with cells plated onto poly-L-ornithine-coated 24-well culture dishes. To evaluate whether modifications in the number of precursors, neurons, or astrocytes were associated with proliferation, cells were double labeled for nestin and BrdU, GFAP and BrdU, and ß-tubulin and BrdU. BrdU (10 µM) was added for 24 h before the end of the experiment. Cells were fixed in 4% paraformaldehyde in PBS for 10 min at 4 C and permeabilized with ETOH/acetic for 20 min at -20 C. After blocking with normal goat serum for 20 min at room temperature, cells were incubated overnight with the corresponding primary antisera at 4 C. Polyclonal antisera used were nestin (1:400 dilution) (M. Vallejo) and GFAP (1:250 dilution) (Sigma). Monoclonal antibody for ß-tubulin isotype III (1:250 dilution) (Sigma) was used.
For immunofluorescence, cells were incubated in fluorescein-conjugated goat antimouse or in rhodamine-conjugated goat antirabbit secondary antisera for 1 h and viewed under epifluorescence optics.
For BrdU immunocytochemistry, cells were fixed with 4% paraformaldehyde in PBS for 15 min, postfixed in 70% ethanol for 30 min, and treated with 2 N HCl for 10 min; blocked with 3% BSA, 5% goat serum, 0.3% Triton in PBS for 1 h; incubated in mouse anti-BrdU 1:20 in blocking solution 0/N at 4 C. For immunofluorescence, cells were incubated in fluorescein-conjugated goat antimouse secondary antibody (1:100) for 1 h and viewed under fluorescein epifluorescence optics.
For double staining with anti-nestin and anti-BrdU antibody, cells were fixed and permeabilized as described above and then incubated with anti-BrdU antibody (1:20), followed by fluorescein-conjugated goat antimouse (1:100) IgG. After incubation with anti-nestin antisera (1:400) overnight at 4 C, cells were incubated with goat antirabbit rhodamine-conjugated secondary antisera (1:100). Similar procedures were used for double staining with anti-GFAP (1:250 dilution) and BrdU and anti-ß-tubulin (1:250) and anti-BrdU. For quantitation of single and double staining, dishes were moved to random locations on the stage of an immunofluorescence microscope and cells within the field of a 0.5-mm2 eyepiece grid were counted. Twenty fields of vision per dish were examined in two different dishes from three independent experiments.
Statistical analysis
Mean and SE were used to describe each of the variables analyzed. Comparison of means in the different groups was done using an analysis of the one-way ANOVA test followed by the Scheffé Ftest using the SATVIEW program. For the least squares regression analysis we used the SPSS version 9.0 software program for windows (SPSS, Inc., Chicago, IL) to determine which of the independent variables (doses of GH and time) had the stronger association with proliferation (dependent variable). For variable selection we used the method "enter" in which all independent variables are entered in a single step. The difference was considered statistically significant when P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
shows the identification of GHR, IGF-IR, and IGF-I mRNAs by RNase protection assay. The expression of IGF-I by cells from E14 was very low and increased with the embryonic age. Thus, these data indicate that fetal cerebral cortical cells from E14 and E17 express the receptors and the signal involved in the action of the GH/IGF-I system.
|
The relationship between proliferation (as dependent variable) and doses of GH (0, 0.5, 5, 50, and 500 ng/ml) and time (as independent variables) are presented in Fig. 2. These data have been evaluated using least squares regression analysis to determine effects of dose of GH, time of incubation with GH, and their interaction. At 24 h, there was no effect of dose on cell proliferation (Rsq = 0.23; Fig. 2A). However, there was a significant effect of dose of GH on cell proliferation at 48 h (Rsq = 0.60, P < 0.001; Fig. 2B) and 72 h (Rsq = 0.64, P < 0.0001; Fig. 2C). The GH action was less effective with lower and higher doses and shorter and longer exposure time; consequently, it would seem that in our experimental model, dose and GH exposure time are critical for GH to promote proliferation.
|
, treatment of fetal cerebral cortical cells with 5 ng/ml of GH for 48 h induced a 2.5-fold increase in PCNA expression (P < 0.01) vs. control. To further confirm the effect of GH on cell proliferation, fetal cerebral cortical cells from E17 were labeled with BrdU (10 µM) 24 h before harvesting, and the number of cells that incorporated BrdU was determined. As shown in Fig. 3B, there was in the presence of GH (5 ng/ml) a striking increase (P < 0.001) in the number of cells that have incorporated BrdU. Cells cultured in the presence of GH (0.5 and 5 ng/ml) for 48 h also had higher incorporation of [3H]-thymidine (Table 1, P < 0.01, P < 0.001 vs. control). With higher doses (50 and 500 ng/ml), the action of GH was less effective.
|
|
. This proliferative effect was observed in cells from both E14 (P < 0.001) and E17 (data not shown) at 3 DIV. However, when the cells from E14 were maintained for 7 DIV, a decreased number of nestin/BrdU and nestin-positive cells was observed in response to GH treatment. On the other hand, an active proliferation of nestin-positive cells was still evident in the controls (Fig. 4A, c and d). We observed similar results in other experiments with cells from E17 at 5 DIV (data not shown).
|
, a and b). In the cultures treated with GH, a higher number of ß-tubulin cells displaying a more elaborate and differentiated morphology with dendritic trees typical of morphologically differentiated neurons was observed (Fig. 4B, b). GH had no proliferative effects on GFAP-positive cells from E14. Under GH treatment, the number of cells identified with double-staining GFAP/BrdU was similar to that found in the controls (Fig. 4B, c and d). Neither the number nor morphology of GFAP-positive cells was modified by GH.
GH induces proliferation of astrocytes in cells from E17
To further evaluate the possible role of GH in regulating the development of astrocytes, cells from E17 were used. GFAP expression was examined in the dividing cells by immunocytochemistry and Western immunoblotting. Cultures growing in the presence of GH (50 ng/ml) for 2 DIV were labeled with BrdU 24 h before harvesting and subsequently a dual-antigen immunocytochemistry with anti-BrdU and anti-GFAP antibodies was performed. Immunofluorescence microscopy of these cells showed a marked increase in GFAP/BrdU-positive cells in GH-treated (50 ng/ml) cultures (P < 0.01). GH also increased the number of cells that expressed GFAP (d) relative to control (c) (Fig. 4C). The cells treated with GH exhibited intense GFAP immunostaining, localized to the cell soma and the processes that displayed a denser morphology and greater length (Fig 4C, d).
Analysis of GFAP protein levels by Western immunoblotting revealed that cells cultured in the presence of GH (50 and 500 ng/ml) for 48 h showed a significant increase in GFAP expression (P < 0.001, P < 0.01 vs. control, Fig. 5B). The highest induction of GFAP by GH occurred at a critical dose of 50 ng/ml.
|
, although GH increased the number of proliferating cells (d) in comparison with control (c), there was no significant modification in the percentage of ß-tubulin/BrdU-positive cells (Fig. 4C). However, GH increased the population of ß-tubulin cells and promoted morphological changes in neurons that showed an increase in the number and length of cellular process with mature neuronal morphology.
Cells cultured in the presence of GH (50 ng/ml) for 48 h also showed an increase in ß-tubulin expression as measured by Western immunoblotting (P < 0.01 vs. control, Fig. 5A).
GH activates the IGF-I, IGF-binding protein-3 (IGFBP3), and IGF-IR system
Action of GH on IGF-I protein expression.
To clarify whether the observed neurotrophic actions of GH on fetal cerebral cortical cells were mediated by the effector IGF-I, our first step was to determine whether IGF-I expression in these cells could be induced by GH. For these experiments, E17 cells were grown with GH for 48 h and subsequently labeled with anti-IGF-I antiserum. Immunofluorescence microscopy of these cultures showed a marked increase in IGF-I-positive cells when cells were treated with GH (50 ng/ml) (data not shown). Quantitation of the number of IGF-I-labeled cells is depicted in Fig. 6A.
|
).
GH induction of IGF-IR phosphorylation
We next evaluated possible GH modulation of IGF-IR protein levels and its activation. The results, shown in Fig. 6C, confirm that IGF-IR ß-subunit is expressed in E17 fetal cerebral cortical cells and its levels are increased by GH, the effect being more evident with the lowest dose of GH (5 ng/ml) used. Maximum phosphorylation of IGF-IRß was observed after approximately 30 min of exposure to GH (5 ng/ml) and was maintained for at least 24 h (Fig. 6C).
Blockade by IGF-I-As of GH action on cell proliferation and differentiation
Concomitant incubation with GH and IGF-I-As was used to confirm that IGF mediates on the effects of GH on proliferation and differentiation. Initial studies were undertaken to establish the concentration of purified IgGs obtained from IGF-I antiserum needed to neutralize the proliferative action of IGF-I.
Alteration in total cell number was established by cell counting and Western immunoblotting of PCNA. As shown in Fig. 7, A and B, the stimulatory effect of GH (5 ng/ml) and IGF-I (10 nM) on cell number and PCNA levels was completely abolished by concomitant incubation with IGF-I antiserum. IgGs from normal rabbit serum (NRS) or anti-IGF-I antiserum did not alter basal cell number or PCNA levels, indicating that, at the concentrations used in the study, they were not toxic.
|
, A and B, the stimulatory effects of GH (50 ng/ml) or IGF-I (10 nM) on GFAP and ß-tubulin expression were completely abolished by concomitant incubation with IGF-I antiserum.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The role of GH in the CNS has been only partially explored. Several findings suggest that the brain may be a target of GH during development (12, 16, 30). To explore the role of GH in regulating brain development, we examined its effects on the proliferation and differentiation of fetal cerebral cortical cells using an in vitro system. Our results indicate that the proliferation of fetal cerebral cortical cells from E17 is positively regulated by GH as assessed by cell counting, PCNA analysis, and incorporation of [3H]-thymidine and BrdU. The action of GH on cerebral cortical cell proliferation, shown in the present study, is indicative of a physiological effect. The greatest effect of GH on proliferation occurred at concentrations of 0.5 and 5 ng/ml, which are within the range of physiological concentrations of GH in the adult rat serum. Higher doses of 50 and 500 ng/ml were less effective. This observation that a critical low dose is needed for GH to exert its proliferative effect is in agreement with previous work by Fuh et al. (32) who demonstrated that GH at low concentrations produces an active GH/GHR complex but at high concentrations saturates the receptor and acts as an antagonist. GH belongs to a family of hormones that includes PRL and many cytokines that exhibit bell-shaped dose response curves when excess ligand prevents dimerization of the receptor (31, 32, 33).
To investigate the phenotype of the proliferating cells, immunocytochemistry studies were performed in fetal cerebral cortical cells from E14 and E17. Cultures from E14, which is at the onset of cortical neurogenesis, rich in neural precursor cells, allow examination of neural and neuronal precursor proliferation and the earlier steps of neuronal differentiation. Cultures from E17 are suitable for investigating events related to gliogenesis. Our results indicate that although after short exposure (3 DIV) GH induces proliferation of neural precursor, prolonged treatment reduces proliferation of neural precursors and expression of nestin. This observation suggests that GH might be inducing differentiation of neural precursors. The proliferative action of GH on cortical neural precursor cells is consistent with the level of expression of GH receptor during development (12).
To determine the role of GH in regulating neurogenesis in the developing nervous system, we examined its effects on the proliferation and differentiation of embryonic cortical cells from E14 and E17. At an early stage of development (E14) neurons are induced to proliferate by GH, showing morphological characteristics of more mature neurons and an increased number of dendrites. In cells from E17, no effect of GH was observed on proliferation of neurons, but an increase in the number of neurons with characteristics of more mature neurons was evident. These morphological changes were accompanied by an increase in the levels of ß-tubulin as measured by Western immunoblotting. Thus, in cultures from E17, we observed an increase in the number of neurons as well as levels of ß-tubulin. These neurons were postmitotic because they did not incorporate BrdU. Our results do not allow us to indicate whether differentiation or survival accounts for the increased number of neurons.
This study supports the notion that in the developing mammalian cortex, GH promotes proliferation, differentiation, and maturation of neurons and confirms results of previous in vivo studies in which a higher neuron-glia index was detected in offspring after GH treatment of pregnant rats by histological estimate (34). Our observations indicate that GH induces astrocyte proliferation, thus confirming previous reports (35). This conclusion is supported by the finding that in cells from E17, but not from E14, addition of GH resulted in an increased number of GFAP/BrdU-positive cells. This indicates that GH is an activator of astrocyte generation. Thus, the increase in GFAP protein in cells from E17 could be due to increased proliferation of astroblasts along with astrocyte differentiation. The state of maturation at which glial precursors show the capacity to respond to GH in vitro parallels the appearance of astroglial phenotype in vivo, which occurs after neurogenesis.
These results show that GH can promote both neuron and astrocyte proliferation and these effects depend upon the age at which cerebral cortical cells are exposed to GH. One important observation of this study is that in our culture conditions, cortical precursors cells isolated at different embryonic stages behave in a manner that mimics the normal process of development (36, 37). In agreement with our results, studies performed by other authors demonstrate that neural precursors from rat embryonic d 14, which corresponds to the peak of neurogenesis in vivo, primarily give rise to neurons and dividing precursor cells (38). Furthermore, in E14 cultures, astrocytes are generated only after several days in vitro. In contrast, E17 precursors that are abundant in the cell clusters give immediate rise to astrocytes (39).
This study demonstrates the role of IGF-I in the neurotrophic actions of GH on cultured cerebrocortical cells. GH-induced cell proliferation was completely reversed by concomitant incubation with IGF-I-As. These observations strongly suggest that the action of GH on cerebral cortical cell proliferation is mediated by locally induced IGF-I, although there may be other growth factors or hormones that are permissive to local effects of IGF-I. In addition, the actions of GH on ß-tubulin and GFAP induction were blocked by IGF-I antiserum, further suggesting that effects of GH on cultured fetal cerebral cortical cells are IGF-I mediated.
Our results demonstrate that GH-induced IGF-I receptor protein expression has a bell-shaped dose-response curve, which mimics its effect on proliferation. Results also indicated an increase in phosphorylation state of the IGF-I receptor because of the increase in IGF-I following GH treatment.
We observed an increase in IGF-I protein in response to GH. The coordinated expression of IGF-I and the GH receptor genes in the developing brain suggests tissue-specific induction of IGF-I by GH during development (12, 16). In the postnatal life, there is evidence that GH influences IGF-I expression in the brain of animals and humans (26, 40, 41). IGF-I mRNA abundance is reduced in the brain of hypophysectomized adult rats, and intracerebral infusion of GH restores IGF-I mRNA to 80% of normal, indicating that GH has a role in modulating brain IGF-I (19). Further comprehension of the interaction between GH and IGF-I in different target tissues, and brain in particular, has been gained from the IGF-I conditional knockout mice studies (42) using the Cre/loxP system in which the role of IGF-I was closely analyzed in individual cell types or developmental stages (43, 44).
Results of the present study also confirm that all the components of GH/IGF-I/IGFBP3 somatotrophic system are expressed in cerebral cortical cells established in culture. Although all cells and tissues that have been examined synthesize at least one form of IGFBP, the general pattern is that each tissue or cell type synthesizes different combinations. Thus, in neural cell types (45, 46), basal expression of IGFBP2 as the major IGFBP along with a small quantity of IGFBP4 has been reported. Also, expression of IGFBP3 has been demonstrated in rat ventral mesencephalic cultures (47). We observed an evident increase in IGFBP3 in response to GH, confirming results of previous studies that IGFBP3 was expressed after GH and IGF-I treatment (48, 49). Also, in the little mouse, the deficiency of GH, coupled with the GH-induced secondary deficiency in IGF-I, resulted in reduced IGFBP3 levels at all ages studied (50).
In this study, several factors may account for the linear increase in IGFBP3 observed in response to GH. Changes in the stability of the IGFBP3/IGF-I complex, and in the affinity of the IGFBP3, or the GH-stimulated expression of IGF-I that was, in turn, bound by IGFBP3, should not be excluded. Also, GH may induce other genes that regulate the expression of IGFBP3. Neither IGFBP2 nor IGFBP4 was detected in our cultures. These findings support a role for IGFBP3 in GH action that is IGF-I dependent by modulating IGF-I availability and activity or by IGF-I-independent mechanisms.
In summary, we have shown that GH at physiological doses is a potent inducer of proliferation of neural cell precursors, neurogenesis, and gliogenesis in fetal cerebral cortical cells. GH actions are fully mediated by IGF-I whose expression is locally induced by GH. The fact that GH actions are mediated throughout the activation of brain IGF-I encourages the possible use of GH as a therapeutic agent. These data also demonstrate that GH exerts effects on fetal cerebral cells, which is more physiological than the effect of IGF-I and therefore potentially useful for human therapy. As a whole, results of the present study provide strong support for an endogenous GH/IGF-I axis involved in brain growth and maturation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: BrdU, Bromodeoxyuridine; CNS, central nervous system; DIV, day in vitro; E14, 14-d-old embryo; E17, 17-d-old embryo; FBS, fetal bovine serum; GFAP, glial fibrillary acidic protein; GHR, GH receptor; HBSS, Hanks’ balanced salt solution; HS, horse serum; IGFBP3, IGF-binding protein-3; IGF-I-As, IGF-I-antiserum; IGF-IR, IGF-I receptor; NRS, normal rabbit serum; PCNA, proliferating cell nuclear antigen; rhGH, recombinant human GH; RNase, ribonuclease.
Received July 1, 2002.
Accepted for publication November 19, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A.-L. Svensson, N. Bucht, M. Hallberg, and F. Nyberg From the Cover: Reversal of opiate-induced apoptosis by human recombinant growth hormone in murine foetus primary hippocampal neuronal cell cultures PNAS, May 20, 2008; 105(20): 7304 - 7308. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. C. Russo, P. D. Gluckman, E. L. Feldman, and G. A. Werther The Insulin-Like Growth Factor System and Its Pleiotropic Functions in Brain Endocr. Rev., December 1, 2005; 26(7): 916 - 943. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Fleenor, J. Oden, P. A. Kelly, S. Mohan, S. Alliouachene, M. Pende, S. Wentz, J. Kerr, and M. Freemark Roles of the Lactogens and Somatogens in Perinatal and Postnatal Metabolism and Growth: Studies of a Novel Mouse Model Combining Lactogen Resistance and Growth Hormone Deficiency Endocrinology, January 1, 2005; 146(1): 103 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. C. Russo, S. Metaxas, K. Kobayashi, M. Harris, and G. A. Werther Antiapoptotic Effects of Leptin in Human Neuroblastoma Cells Endocrinology, September 1, 2004; 145(9): 4103 - 4112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
